On 2011-05-03 01:45, Michael Sevakis wrote: Playback stopped again, and I managed to write down some information:
Could be that I regressed some part of the buffering that comes into play more often with smaller memory with Fuze. What Buschel is describing in IRC could be related, but maybe not. I would like to know from each person experiencing this: 1) Near a track transition (say, the last 2-3 seconds of a track if not using crossfade) or not? Which audio formats are involved?
At 4:58 of 5:46. MP3, 64 kbps.
2) What are the thread states as reported by the OS stacks screen when things are stuck? (This helps tell just who is wrong). Codec with "B" while playing is bad and only should be present in a stopped state; it should be "*R" or flipping "*R"/"T" really fast - be sure to change refresh timing by holding down an inactive key or it can appears the codec threads is in one state but is really flipping between a couple.
Codec: *R 16 16 Audio: T 15 15 Buffering: T 15 15 Didn't check for flipping... Also, the buffering debug screen had this to say: pcm: 0 real: ~90k usefl: ~8k pcmbufdesc: 0 track count: 3 -- Magnus
